Nonlinear Opt.: Gold section search 1: Unterschied zwischen den Versionen
Aus Operations-Research-Wiki
| [unmarkierte Version] | [unmarkierte Version] |
Iacob (Diskussion | Beiträge) (Die Seite wurde neu angelegt: „Interval reduction methods usually use the function value of two interior points in the interval to decide the direction in which to reduce it. One elegant way…“) |
Iacob (Diskussion | Beiträge) |
||
| Zeile 1: | Zeile 1: | ||
| − | Interval reduction methods usually use the function value of two interior | + | Interval reduction methods usually use the function value of two interior points in the interval to decide the direction in which to reduce it. One elegant way is to recycle one of the evaluated points and to use it in the nextiterations. This can be done by using the so-called '''Golden Section rule'''.This method uses two evaluated points l (left) and r (right) in the interval [ak, bk], that are located in such a way that one of the points can be used again in the next iteration. The idea is sketched in Figure 5.1. The evaluation pointsl and r are located with fraction τ in such a way that l = a+(1−τ )(b−a) and r = a + τ (b − a). |
| − | points in the interval to decide the direction in which to reduce it. One | + | |
| − | elegant way is to recycle one of the evaluated points and to use it in the | + | |
| − | + | ||
| − | This method uses two evaluated points l (left) and r (right) in the interval | + | |
| − | [ak, bk], that are located in such a way that one of the points can be used again | + | |
| − | in the next iteration. The idea is sketched in Figure 5.1. The evaluation | + | |
| − | + | ||
| − | r = a + τ (b − a). | + | |
Version vom 11. Juni 2013, 17:10 Uhr
Interval reduction methods usually use the function value of two interior points in the interval to decide the direction in which to reduce it. One elegant way is to recycle one of the evaluated points and to use it in the nextiterations. This can be done by using the so-called Golden Section rule.This method uses two evaluated points l (left) and r (right) in the interval [ak, bk], that are located in such a way that one of the points can be used again in the next iteration. The idea is sketched in Figure 5.1. The evaluation pointsl and r are located with fraction τ in such a way that l = a+(1−τ )(b−a) and r = a + τ (b − a).
